The meaning of carbonate Zn isotope records: Constraints from a detailed geochemical and isotope study of bulk deep-sea carbonates

Studies of the zinc isotope (δ66Zn) composition of the past ocean often exploit the bulk carbonate sediment as an archive. The relationship of bulk carbonate δ66Zn to seawater values, however, is poorly understood. Experimental results suggest that inorganic carbonate records δ66Zn that is substantially higher than the aqueous Zn2+ pool. On the other hand, biogenic carbonate from modern cold-water corals and benthic foraminifera appear to record the ambient seawater δ66Zn value, without fractionation. Here, we address this problem through a detailed study of the different geochemical and isotope reservoirs in deep-sea carbonate samples. Carbonate-rich, pelagic sediment samples were used in this study, one of Holocene age and one from the mid-Cretaceous. The samples were subjected to extensive leaching experiments and interpreted using a mixing model to identify silicate and Fe-Mn oxide contaminants as well as pure carbonate phases. The chemical and isotopic characteristics of the contaminant phases are consistent with previous analyses of silicates and Fe-Mn oxides. The δ66Zn values of the pure carbonate end-members, identified using a mixing model, at +0.93 to +1.10‰, are within analytical uncertainty of results obtained via leaching with buffered 1 M acetic acid (pH 5). By contrast, the δ66Zn of un-buffered acetic acid leachates are up to 0.35‰ beneath these values, reflecting the influence of contaminant phases. Weak mineral acids (e.g. 0.2–0.5 M HCl) are lighter still, at up to 0.5‰ below carbonate end-members. The bulk carbonate sediment end-member δ66Zn for the Holocene sample are ∼0.37–0.54‰ above deep ocean values, consistent with the experimentally determined isotope offset between inorganic carbonate precipitates and an aqueous fluid. These carbonate end-members do not appear to directly record seawater compositions, unlike some biogenic carbonates. These differences identify the potential for ancient sediment δ66Zn records to reflect variations in Zn incorporation mechanism into the carbonate, rather than the δ66Zn of contemporary seawater, and hence may not reflect changes in Zn sources and sinks driven by environmental change. Further research is required to better identify the controls of δ66Zn in bulk carbonate sediments to reliably use them as an archive for seawater.